We are expanding our GPU server lineup to offer dedicated servers built around NVIDIA RTX A5000 and RTX A4000 GPU cards. The servers are located in a data center in the Netherlands.
We pride ourselves on our ability to provide our customers with new high-performance GPU servers with the latest cards and the best price-performance ratio.
NVIDIA RTX A5000 and RTX A4000 cards are built on the latest Ampere architecture and are, therefore, ideal for projects in the field of 3D Graphics, Artificial Intelligence, Digital Information Visualization, Cyber Security - all projects wherein a large amount of parallel computing is required.
Basic server configurations and prices will be announced soon.
Contact us, if you would like to pre-order a GPU server based on these new cards now. The demand for GPU servers is very high, and their number is limited.
NVIDIA launched its RTX A5000 and RTX A4000 graphics cards on April 12th, 2021. These new cards are the closest relatives of the RTX 3080 and RTX 3070, but the RTX A5000 and RTX A4000 have double the memory.
The RTX A5000 is based on the GPU GA102. It has 8,192 CUDA cores, 256 Tensor cores and 64 Ray Tracing cores, 24GB of memory with a 384-bit bus. Its peak memory bandwidth is claimed to be 768 GB/s. According to NVIDIA, the RTX A5000 delivers 27.8 teraflops of single precision performance. The performance of RT cores in ray tracing tasks can reach 54.2 teraflops, and the performance of tensor cores can reach 222.2 teraflops.
The RTX A4000 is built upon the GPU GA104. It has 6,144 CUDA cores, 192 tensor cores and 48 Ray Tracing cores. The video card is equipped with 16 GB of GDDR6 video memory with a 256-bit bus and an ECC error correction function. The theoretical memory bandwidth is 448 GB/s. The peak performance of the RTX A4000 in single precision operations reaches 19.2 teraflops, in ray tracing tasks - up to 37.4 teraflops, and the performance of its tensor cores can reach 153.4 teraflops.